Lyft Data Science interview questions

1) Lyft wants to launch ride share program. What are the most important KPIs would you track? 2) Lyft is trying to optimize ride sharing program. The app shows time for a ride to arrive. How would you decide a time threshold before customer cancels the request? 3) A $5 coupon is given to N riders. Probability of using coupon is P. What is expected spending? 4) A $5 coupon is given to N riders. Probability of using coupon is P. What is the probability that one of the coupons will be used? 5) There are 2 riders. Both have coupon. Given that at least one of them uses a coupon, what is the probability that both uses the coupon? 6) In the above question we assumed that the events of using the coupon are independent. In what case our assumption would be incorrect? 7) Currently, lyft app shows estimated time for a ride to arrive (ex. estimated time of arrival 5 min). We are thinking to change it to show a range (ex, estimated time of arrival 3-5 min). Hypothesis is that total rides would increase. How would you go about setting up the experiment? 8) Our internal dashboard shows that number of new drivers is down by 7%. How would you investigate this issue? 9) Is it necessary to investigate decrease in new drivers signed up? In what case it will not be necessary? 10) I do not understand P value. Will you explain it?

How do you explain p value to someone who's not technical?